Wearable body area network and sensors for healthcare applications

Abstract

With an increased demand of wearable healthcare systems, a low power body area network (BAN) becomes much more important for efficient and effective operation of the system. Among several wearable healthcare applications, sleep monitoring system should become much more convenient than other wearable healthcare systems since it always operates during sleep which is the uncontrolled state and its electrodes are placed on the very small area of the face. For maximum convenience, the following system requirements can be derived: 1) The electrodes should be as small as possible, 2) The wires outsides of the sleeper should be removed as much as possible, and 3) The weight of the total system should be as light as possible. In addition, compensation techniques are essential for the dynamic environment due to sleeper’s tossing and turning during sleep. This dissertation proposes a convenient sleep monitoring system that can deal with the whole above requirements. It consists of 3 low power SoCs for wearable BAN, and is implemented on the wearable circuit board technology, named Planar Fashionable Circuit Board (P-FCB). The first SoC is an ExG sensor. It includes 2 capacitive feedback instrumentation amplifiers for enough signal amplification and 10-bit SAR ADC for low power digitization. To deal with 4 different kinds of ExG signals including EEG, EOG, EMG, and ECG, it can control the amp gain and bandwidth. The sensing data is transmitted through the data transceiver in the SoC. It communicates with the WBAN controller chip with low power consumption. It adopts both low swing data transmission and duty-cycled operation so that it consumes only 0.33pJ/b with 20Mbps data rate in average. This sensor SoC is implemented in 0.18um CMOS process, and dissipates 25uW with 1.5V supply voltage. The second SoC is a WBAN controller which provides the real-time sensor network management of joining and leaving with low power consumption. For energy efficient network management, t...